Hmm alright, I shall poke the archives and see what useful information i can obtain. =)
what you describe is pretty much what I have been doing, playing with the data for just the .15 2Kv caps in an excel spread sheet to nail all the formulas down. once i am done that i will expand it to the other cases in an actual programming sense.
I have made several poly caps previously, and to date none of them have failed. Two open air units and one under oil.
I got around the need for thick Aluminium sheeting by tapping each turn of the roll with a nice long strip of aluminium and tying them all together, this allows many paths for the charge to escape. Minimizing the over all current flowing threw any given junction.Thus minimizing heating and the like. This also means for a pretty low inductance.
I have run these guys for very long runs, and they get only lightly warm to the touch and while being over driven for what I designed them for as well. Work for me.
Thus far solely in terms of cost analysis, the poly caps are by far the cheapest route to go for me. Due to maxwells being so bloody expensive , and the sheer quantity of MMC caps i need to get the RMS currents to work out. That's with my RMS current equation thought, which I am trying to improve upon. Best case scenario is as I make my equation more accurate, the RMS current drops and I need less capacitors. or I stumble upon some very useful information regarding the ratios of actual RMS current to rated RMS current. Which from what I found last night, is information that does exist somewhere, I just need to dig it up again.
Speaking of maxwells, what is the rms current that these bad boys can handle for short runs? I have a 0.15uf 50Kv unit I acquired at a steal price. says 25A rms.
What do you guys run them at?